A study on autonomous learning mechanism of Cognitive robot

Aiming at the movement balance control problem of the robot, this paper presents a sensorimotor system Cognitive model based on operant conditioning principle, and researches the working cooperation among its interior nerve organs, so that a sensorimotor system is established. The Cognitive model can realize the sensorimotor mapping from states to actions by supervised learning, and carry out the probabilistic choice based on operant conditioning principle to actions using the action forecast evaluation results, thereby, the robot obtains the self-learning ability like human or animal through interacting, studying and training with the unknown environment, and realized the movement balance control to the robot. Consequently, the paper makes some simulation experiments on the robot, and the results indicate that this model has the better Cognitive characters and make the robot master the movement balance control skill through autonomic learning.